The n e w e ng l a n d j o u r na l of m e dic i n e

Original Article

Encorafenib, Binimetinib, and Cetuximab

in BRAF V600E–Mutated Colorectal Cancer
S. Kopetz, A. Grothey, R. Yaeger, E. Van Cutsem, J. Desai, T. Yoshino, H. Wasan,
F. Ciardiello, F. Loupakis, Y.S. Hong, N. Steeghs, T.K. Guren, H.-T. Arkenau,
P. Garcia‑Alfonso, P. Pfeiffer, S. Orlov, S. Lonardi, E. Elez, T.-W. Kim,
J.H.M. Schellens, C. Guo, A. Krishnan, J. Dekervel, V. Morris, A. Calvo Ferrandiz,
L.S. Tarpgaard, M. Braun, A. Gollerkeri, C. Keir, K. Maharry, M. Pickard,
J. Christy‑Bittel, L. Anderson, V. Sandor, and J. Tabernero​​

A BS T R AC T

BACKGROUND
The authors’ full names, academic de- Patients with metastatic colorectal cancer with the BRAF V600E mutation have a
grees, and affiliations are listed in the poor prognosis, with a median overall survival of 4 to 6 months after failure of
Appendix. Address reprint requests to
Dr. Tabernero at Vall d’Hebron University initial therapy. Inhibition of BRAF alone has limited activity because of pathway
Hospital, Vall d’Hebron Institute of On- reactivation through epidermal growth factor receptor signaling.
cology (VHIO), Passeig de la Vall d’Hebron
119-129, 08035 Barcelona Spain, or at METHODS
­jtabernero@​­vhio​.­net.
In this open-label, phase 3 trial, we enrolled 665 patients with BRAF V600E–
A list of committee members and princi- mutated metastatic colorectal cancer who had had disease progression after one or
pal investigators in this trial is provided in
the Supplementary Appendix, available
two previous regimens. Patients were randomly assigned in a 1:1:1 ratio to receive
at NEJM.org. encorafenib, binimetinib, and cetuximab (triplet-therapy group); encorafenib and
This article was published on September
cetuximab (doublet-therapy group); or the investigators’ choice of either cetuximab
30, 2019, at NEJM.org. and irinotecan or cetuximab and FOLFIRI (folinic acid, fluorouracil, and irinotecan)
N Engl J Med 2019;381:1632-43.
(control group). The primary end points were overall survival and objective response
DOI: 10.1056/NEJMoa1908075 rate in the triplet-therapy group as compared with the control group. A secondary
Copyright © 2019 Massachusetts Medical Society. end point was overall survival in the doublet-therapy group as compared with the
control group. We report here the results of a prespecified interim analysis.
RESULTS
The median overall survival was 9.0 months in the triplet-therapy group and 5.4
months in the control group (hazard ratio for death, 0.52; 95% confidence interval
[CI], 0.39 to 0.70; P<0.001). The confirmed response rate was 26% (95% CI, 18 to
35) in the triplet-therapy group and 2% (95% CI, 0 to 7) in the control group
(P<0.001). The median overall survival in the doublet-therapy group was 8.4 months
(hazard ratio for death vs. control, 0.60; 95% CI, 0.45 to 0.79; P<0.001). Adverse
events of grade 3 or higher occurred in 58% of patients in the triplet-therapy
group, in 50% in the doublet-therapy group, and in 61% in the control group.
CONCLUSIONS
A combination of encorafenib, cetuximab, and binimetinib resulted in signifi-
cantly longer overall survival and a higher response rate than standard therapy in
patients with metastatic colorectal cancer with the BRAF V600E mutation. (Funded
by Array BioPharma and others; BEACON CRC ClinicalTrials.gov number,
NCT02928224; EudraCT number, 2015​-­005805​-­35.)

1632 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

T
he BRAF V600E mutation occurs in when compared with reported outcomes with
approximately 10% of patients with meta- standard treatment options and outcomes from
static colorectal cancer, with recent esti- early-phase trials of triplet regimens that com-
mates ranging from as low as 5% to as high as bined an anti-EGFR antibody with other BRAF
21%.1-10 This mutation identifies a distinct sub- inhibitors plus either an MEK inhibitor or irino-
type of colorectal cancer that has a poor prog- tecan.13,21,27 We therefore sought to evaluate
nosis.11 Initial standard chemotherapy for BRAF whether treatment with the combination of en-
V600E–mutated colorectal cancer results in poor corafenib plus cetuximab with or without the
outcomes, and attempts to intensify therapy have MEK inhibitor binimetinib would result in longer
met with limited success.12 After failure of initial overall survival than standard therapy in patients
therapy, subsequent lines of treatment have with BRAF V600E–mutated metastatic colorectal
minimal effect, and rapidly progressive disease cancer who had had disease progression after
and short overall survival are seen. Recently re- one or two previous lines of therapy. Before ini-
ported results of a phase 2 trial of second-line tiation of the randomized phase of the trial, a
and third-line therapy in this population showed group of patients was enrolled in a lead-in phase
an objective response rate of 4%, progression- to evaluate the safety of the triplet regimen.28
free survival of 2 months, and overall survival of
5.9 months in the control group, which received Me thods
standard therapy (irinotecan plus cetuximab).13
Although BRAF V600E is a driver mutation Trial Oversight
found in multiple tumor types, and BRAF in- The trial was approved by the institutional re-
hibitors have clinical activity in BRAF V600E– view board or independent ethics committee at
mutated melanoma and non–small-cell lung can- each center and was conducted in accordance
cer, BRAF inhibitors alone have limited activity with the requirements of the regulatory authori-
in BRAF V600E–mutated colorectal cancer.14-17 Pre- ties of each country and with the provisions of
clinical models of BRAF V600E–mutated colorec- the Declaration of Helsinki and the Good Clini-
tal cancer have shown that BRAF inhibition cal Practice guidelines of the International Coun-
causes rapid feedback activation through the epi- cil on Harmonisation. All patients provided
dermal growth factor receptor (EGFR); thus, written informed consent. The steering commit-
BRAF inhibitor treatment alone does not suffi- tee and one of the sponsors (Array BioPharma)
ciently inhibit pathway signaling, which explains jointly designed the trial and reviewed the data,
the lack of clinical efficacy of BRAF inhibition with the participation of the authors. All the in-
in this type of cancer.18-20 Translation of these vestigators collected the data. Six of the authors
findings into clinical trials has shown that BRAF wrote the first draft of the manuscript, with
inhibitors have increased antitumor activity in professional medical writing assistance funded
BRAF V600E–mutated colorectal cancer when com- by Array BioPharma, and all the authors contrib-
bined with anti-EGFR monoclonal antibodies.21-23 uted to subsequent drafts. All the authors vouch
Preclinical studies have suggested that combined for the accuracy and completeness of the data
inhibition of BRAF and mitogen-activated pro- and analyses and for the fidelity of the trial to
tein kinase (MAPK) kinase (MEK) is more effec- the protocol (available with the full text of this
tive than BRAF inhibitors combined with anti- article at NEJM.org).
EGFR agents. This finding was validated clinically
in subsequent phase 1 and phase 2 trials that Trial Design, Patients, and Treatments
combined BRAF inhibitors with both anti-EGFR The BEACON CRC (Binimetinib, Encorafenib,
monoclonal antibodies and MEK inhibitors.18,24,25 and Cetuximab Combined to Treat BRAF-Mutant
Encorafenib is a BRAF inhibitor with more Colorectal Cancer) trial is a global, multicenter,
prolonged pharmacodynamic activity than other randomized, open-label, phase 3 trial. We en-
approved BRAF inhibitors.26 The combination of rolled patients with histologically or cytologically
the BRAF inhibitor encorafenib and the anti- confirmed metastatic colorectal cancer with the
EGFR monoclonal antibody cetuximab showed BRAF V600E mutation who had had disease pro-
promising activity in early-phase clinical trials gression after one or two previous treatment

n engl j med 381;17 nejm.org October 24, 2019 1633

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

regimens. Additional inclusion and exclusion vival, duration of response, and safety in all
criteria are provided in the trial protocol. groups. Tumor assessments were performed ac-
We randomly assigned patients in a 1:1:1 ratio cording to Response Evaluation Criteria in Solid
to one of three groups. Patients in the triplet- Tumors (RECIST), version 1.1,29 every 6 weeks
therapy group received encorafenib (300 mg daily), from the date of randomization for the first 24
binimetinib (45 mg twice daily), and cetuximab weeks of treatment, then every 12 weeks there-
(400 mg per square meter of body-surface area after until disease progression, withdrawal of
as an initial dose, then 250 mg per square meter consent, initiation of subsequent anticancer
weekly). Patients in the doublet-therapy group therapy, loss to follow-up, or death, regardless
received encorafenib and cetuximab, adminis- of whether the trial treatment was discontinued.
tered in the same doses and on the same sched- All responses were confirmed by means of sub-
ule as the triplet regimen. Patients in the control sequent imaging performed at least 4 weeks after
group received the investigators’ choice of either the initial response. The central review of imag-
cetuximab (administered in the same doses and ing data was performed retrospectively by read-
on the same schedule as the other regimens) and ers who were unaware of the treatment assign-
irinotecan (180 mg per square meter on days 1 ments. The incidence and severity of adverse
and 15) or cetuximab and FOLFIRI (folinic acid events were assessed according to the National
[180 mg per square meter, administered on days Cancer Institute Common Terminology Criteria
1 and 15], fluorouracil [400 mg per square meter for Adverse Events, version 4.03.30
as an initial dose, then 1200 mg per square meter
per day for 2 days, initiated on days 1 and 15], Statistical Analysis
and irinotecan [at the same dose and on the The type I error rate for the primary end points
same schedule as the other regimens]). Random- was controlled with the use of a fallback proce-
ization was stratified according to Eastern Co- dure described by Wiens and Dmitrienko.31 The
operative Oncology Group (ECOG) performance- end point of objective response rate in the trip-
status score (0 or 1 [scores range from 0 to 5, let-therapy group as compared with the control
with higher scores reflecting greater disability]), group was assigned a one-sided alpha level of
previous use of irinotecan (yes or no), and cetux- 0.005. The remaining 0.020 was assigned to the
imab formulation (U.S.-licensed or European- end point of overall survival in the triplet-therapy
approved formulation). Treatment was adminis- group as compared with the control group.
tered in 28-day cycles until disease progression, To incorporate testing of selected secondary
unacceptable toxic effects, withdrawal of consent, end points, a gatekeeping procedure with hierar-
initiation of subsequent anticancer therapy, or chical testing was also used to account for the
death. Crossover was not permitted before the multiple comparisons (see the Supplementary
data cutoff date. Appendix, available at NEJM.org). The trial was
not powered to formally compare the results in
End Points the triplet-therapy group with those in the dou-
The original sole primary end point was overall blet-therapy group.
survival in the triplet-therapy group as compared The sample size was driven by the secondary
with the control group. The protocol was amend- end point of overall survival in the doublet-ther-
ed to include an additional primary end point of apy group as compared with the control group.
the objective response rate in the triplet-therapy For this comparison, we calculated that 338
group as compared with the control group, as deaths would be required to give the trial 90%
assessed by independent central reviewers who power to detect a hazard ratio for death of 0.70,
were unaware of the treatment assignments, and with the use of a stratified log-rank test at a one-
the initial interim analysis of overall survival (the sided significance level of 0.025. The number of
results of which are reported here) was added in patients who would need to be included in the
an attempt to expeditiously assess efficacy. Sec- primary analysis of objective response rate in the
ondary end points included overall survival in triplet-therapy group as compared with the con-
the doublet-therapy group as compared with the trol group was based on an assumption that the
control group, as well as progression-free sur- objective response rate would be 10% in the

1634 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

control group and 30% in the triplet-therapy interval [CI], 8.0 to 11.4) in the triplet-therapy
group; we calculated that 110 patients per group group and 5.4 months (95% CI, 4.8 to 6.6) in the
would provide 88% power, at a one-sided signifi- control group. The risk of death was significantly
cance level of 0.005, to show the higher objective lower (by 48%) in the triplet-therapy group than
response rate in the triplet-therapy group. The in the control group (hazard ratio, 0.52; 95% CI,
end point was tested with the use of a Cochran– 0.39 to 0.70; P<0.001). The results of subgroup
Mantel–Haenszel test. The initial analysis was analyses were consistent with those of the over-
performed by an independent statistician and all analysis (Fig. 1C). The median overall sur-
reviewed by the independent data monitoring vival was 8.4 months (95% CI, 7.5 to 11.0) in the
committee, which recommended whether the doublet-therapy group, and the risk of death was
sponsor would have access to unblinded data on significantly lower than that in the control
the basis of planned boundaries for futility and group (hazard ratio, 0.60; 95% CI, 0.45 to 0.79;
superiority. P<0.001). In a descriptive analysis of survival
Unless otherwise stated, the time-to-event end comparing the triplet regimen with the doublet
points were assessed in all patients who under- regimen, the estimated 6-month survival was
went randomization (i.e., the intention-to-treat 71% in the triplet-therapy group and 65% in the
population). Objective response was assessed in doublet-therapy group (hazard ratio for death,
the first 331 patients who underwent random- 0.79; 95% CI, 0.59 to 1.06) (Fig. S2). In the control
ization. Safety was assessed on the basis of ad- group, the estimated 6-month survival was 47%.
verse events and laboratory abnormalities and The objective response rate was significantly
was evaluated in patients who received at least higher in the triplet-therapy group than in the
one dose of any trial drug and had at least one control group (Table 2 and Fig. 2). The indepen-
post-treatment safety assessment. For the pur- dently reviewed confirmed objective response rate,
poses of this summary, all reported P values are assessed in the first 331 patients who underwent
two-sided. Additional details regarding the trial randomization, was 26% (95% CI, 18 to 35) in
design and analysis methods are provided in the the triplet-therapy group and 2% (95% CI, 0 to 7)
Supplementary Appendix. in the control group (P<0.001). The objective
response rate in the doublet-therapy group was
R e sult s 20% (95% CI, 13 to 29), which was also signifi-
cantly higher than that in the control group
Patients (P<0.001). A waterfall plot showing the best
A total of 1677 patients were screened for eligi- percentage change from baseline in the sum of
bility. From May 2017 through January 2019, a the diameters of the target lesions in the triplet-
total of 665 patients underwent randomization: therapy group and the doublet-therapy group, as
224 patients were assigned to the triplet-therapy assessed by central review, indicates differences
group, 220 patients to the doublet-therapy group, in the depth of response that favor the triplet
and 221 patients to the control group (Fig. S1 in regimen; the responses as assessed by local in-
the Supplementary Appendix). The characteristics vestigators were similar to those assessed by
of the patients at baseline were similar across central review (Fig. S3 and Table S2). The influ-
the three groups (Table 1). Baseline characteris- ence of the number of lines of previous therapy
tics of the 331 patients included in the primary on response rate was assessed on the basis of
analysis of objective response rate are described observations of higher response rates in earlier
in Table S1. lines of therapy in phase 2 studies and observa-
tions in the safety lead-in phase.28,32,33 The re-
Efficacy sponse rate among patients with only one previ-
As of the data cutoff date for this prespecified ous line of therapy was 34% (95% CI, 23 to 47)
interim analysis (February 11, 2019), the median in the triplet-therapy group, 22% (95% CI, 14 to
duration of follow-up for survival was 7.8 months 33) in the doublet-therapy group, and 2% (95%
across the three groups. Kaplan–Meier curves CI, 0 to 9) in the control group.
for survival are shown in Figure 1. The median Progression-free survival (as assessed by cen-
overall survival was 9.0 months (95% confidence tral review) was significantly longer in both the

n engl j med 381;17 nejm.org October 24, 2019 1635

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Baseline Characteristics of the Patients Who Underwent Randomization.*

Triplet Regimen Doublet Regimen Control

Characteristic (N = 224) (N = 220) (N = 221)
Sex — no. (%)
Male 105 (47) 115 (52) 94 (43)
Female 119 (53) 105 (48) 127 (57)
Age — yr
Median 62 61 60
Range 26–85 30–91 27–91
ECOG performance-status score — no. (%)†
0 116 (52) 112 (51) 108 (49)
1 108 (48) 104 (47) 113 (51)
2 0 4 (2) 0
Location of primary tumor — no. (%)
Left side of the colon, including rectum 79 (35) 83 (38) 68 (31)
Right side of the colon 126 (56) 110 (50) 119 (54)
Both left and right side of the colon or unknown location 19 (8) 27 (12) 34 (15)
Involvement of ≥3 organs — no. (%) 110 (49) 103 (47) 98 (44)
Presence of liver metastases — no. (%) 144 (64) 134 (61) 128 (58)
Primary tumor removed — no. (%)
Completely resected 133 (59) 123 (56) 122 (55)
Partially resected or unresected 91 (41) 97 (44) 99 (45)
Previous lines of therapy — no. (%)
1 146 (65) 146 (66) 145 (66)
2‡ 78 (35) 74 (34) 76 (34)
High microsatellite instability — no. (%)§ 22 (10) 19 (9) 12 (5)
Baseline carcinoembryonic antigen level >5 μg/liter — no. (%) 179 (80) 153 (70) 178 (81)
Baseline C-reactive protein level >10 mg/liter — no. (%) 95 (42) 79 (36) 90 (41)

* Percentages may not total 100 because of rounding. The triplet-therapy group received encorafenib, binimetinib, and
cetuximab; the doublet-therapy group received encorafenib and cetuximab; and the control group received the investi-
gators’ choice of either cetuximab and irinotecan or cetuximab and FOLFIRI (folinic acid, fluorouracil, and irinotecan).
† Eastern Cooperative Oncology Group (ECOG) performance-status scores range from 0 to 5, with higher scores indicat-
ing greater disability.
‡ This category includes one patient in the triplet-therapy group and one patient in the control group who received more
than two previous lines of therapy.
§ High microsatellite instability was determined by polymerase chain reaction.

triplet-therapy group and the doublet-therapy

group than in the control group. The median Figure 1 (facing page). Overall Survival.
progression-free survival was 4.3 months (95% Panel A shows the Kaplan–Meier analysis of the proba-
bility of survival in the triplet-therapy group as compared
CI, 4.1 to 5.2) in the triplet-therapy group, 4.2 with the control group, and Panel B the probability
months (95% CI, 3.7 to 5.4) in the doublet-ther- of survival in the doublet-therapy group as compared
apy group, and 1.5 months (95% CI, 1.5 to 1.7) with the control group. Panel C shows the results of
in the control group (Fig. S4). The hazard ratio the subgroup analysis of overall survival in the triplet-
for disease progression or death was 0.38 (95% therapy group as compared with the control group.
Eastern Cooperative Oncology Group (ECOG) perfor-
CI, 0.29 to 0.49) in the triplet-therapy group as mance-status scores range from 0 to 5, with higher
compared with the control group (P<0.001) and scores indicating greater disability.
0.40 (95% CI, 0.31 to 0.52) in the doublet-therapy

1636 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

A Overall Survival, Triplet Regimen vs. Control B Overall Survival, Doublet Regimen vs. Control
Median Overall Survival Median Overall Survival
1.0 mo (95% CI) 1.0 mo (95% CI)
0.9 Triplet 9.0 (8.0–11.4) 0.9 Doublet 8.4 (7.5–11.0)
0.8 Control 5.4 (4.8–6.6) 0.8 Control 5.4 (4.8–6.6)
Probability of Survival

Probability of Survival
0.7 Hazard ratio for death, 0.7 Hazard ratio for death,
0.52 (95% CI, 0.39–0.70) 0.60 (95% CI, 0.45–0.79)
0.6 0.6
P<0.001 P<0.001
0.5 0.5
0.4 0.4
Triplet
0.3 0.3 Doublet
0.2 0.2
Control Control
0.1 0.1
0.0 0.0
0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
Months Months
No. at Risk No. at Risk
Triplet 224 186 141 103 69 37 24 14 6 4 2 0 Doublet 220 184 133 87 57 33 21 12 8 3 1 0
Control 221 158 102 60 34 18 15 7 4 2 1 0 Control 221 158 102 60 34 18 15 7 4 2 1 0

C Overall Survival According to Subgroup, Triplet Regimen vs. Control

Hazard Ratio for Death vs. Control (95% CI)
Subgroup No. of Deaths/No. of Patients
All patients 204/445 0.52 (0.39–0.70)
ECOG performance-status score
0 85/227 0.63 (0.41–0.96)
1 119/218 0.48 (0.33–0.70)
Previous irinotecan use
No 99/219 0.53 (0.35–0.79)
Yes 105/226 0.55 (0.37–0.80)
Region
North America 32/59 0.91 (0.45–1.86)
Europe 124/275 0.39 (0.27–0.56)
Rest of the world 48/111 0.74 (0.41–1.33)
No. of previous regimens for metastatic disease
1 123/291 0.54 (0.38–0.77)
≥2 81/154 0.53 (0.34–0.82)
Age
<65 yr 130/290 0.58 (0.41–0.82)
≥65 yr 74/155 0.48 (0.30–0.76)
Sex
Male 100/199 0.53 (0.36–0.79)
Female 104/246 0.54 (0.36–0.80)
No. of organs involved
≤2 104/237 0.54 (0.37–0.80)
≥3 100/208 0.50 (0.34–0.75)
Microsatellite instability status
Abnormal, high 18/34 0.67 (0.26–1.76)
Normal 140/200 0.44 (0.31–0.62)
Unknown 46/111 0.78 (0.44–1.41)
Baseline carcinoembryonic antigen level
Higher than the upper limit of the normal range 180/257 0.54 (0.40–0.72)
At or below the upper limit of the normal range 23/87 0.42 (0.18–0.99)
Baseline C-reactive protein level
Higher than the upper limit of the normal range 115/185 0.61 (0.42–0.88)
At or below the upper limit of the normal range 83/247 0.46 (0.29–0.71)
Location of tumor
Left side of the colon 53/147 0.43 (0.26–0.72)
Right side of the colon 117/245 0.63 (0.44–0.90)
Both sides of the colon 14/30 0.74 (0.22–2.42)
Unknown 10/23 0.35 (0.09–1.38)
0.1 0.2 0.5 1.0 2.0

Triplet Better Control Better

n engl j med 381;17 nejm.org October 24, 2019 1637

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Tumor Response in Patients with Metastatic Colorectal Cancer with the BRAF V600E Mutation.*

Triplet Doublet
Regimen Regimen Control
Variable (N = 111) (N = 113) (N = 107)
Objective response
Patients with a complete or partial response — no. (%) 29 (26) 23 (20) 2 (2)
95% CI 18–35 13–29 <1–7
P value vs. control <0.001 <0.001
Best overall response — no. (%)
Complete response 4 (4) 6 (5) 0
Partial response 25 (23) 17 (15) 2 (2)
Stable disease† 47 (42) 61 (54) 31 (29)
Progressive disease 11 (10) 8 (7) 36 (34)
Could not be evaluated according to RECIST‡ 24 (22) 21 (19) 38 (36)
Clinical progression or discontinuation because 15 (14) 19 (17) 17 (16)
of adverse event§
Insufficient data to assess response¶ 9 (8) 2 (2) 21 (20)
Patients with duration of response ≥6 mo — no./total no. 7/29 (24) 10/23 (43) 1/2 (50)
of patients with a response (%)
Patients with ongoing response and <6 mo follow-up 4/29 (14) 1/23 (4) 0
— no./total no. of patients with a response (%)

* All responses were confirmed and were assessed by blinded independent central review according to Response Evalu­
ation Criteria in Solid Tumors (RECIST), version 1.1. The first 331 patients who underwent randomization were includ-
ed in the assessment of tumor response. Percentages may not total 100 because of rounding.
† This category refers to patients with measurable disease who had stable disease and patients with nonmeasurable dis-
ease who did not have a complete response or who did not have progressive disease according to RECIST.
‡ This category includes 3 patients in the triplet-therapy group who had a confirmed partial response as determined by
local assessment (these patients underwent scanning at outside institutions at baseline, and the scans were not avail-
able for central assessment).
§ This category refers to patients who discontinued the trial regimen because of adverse events or whose disease could
not be assessed centrally but who had clinical or radiologic disease progression according to local assessment.
¶ This category refers to patients who did not receive treatment, who withdrew consent, whose best response was stable
disease within 42 days after randomization, who had no baseline scans available, or who had no postbaseline scans
and no evidence of clinical progression or an adverse event.

group as compared with the control group trial, grouped according to clinically similar
(P<0.001). As of the data cutoff date, the median events that are commonly associated with BRAF
follow-up for progression-free survival was 5.4 and MEK inhibitors. Class-related toxic effects
months. of MEK inhibitors, including serous retinopathy
and left ventricular dysfunction, occurred at rates
Safety similar to those described previously15,16,28 and
The most frequently occurring adverse events were managed with treatment interruptions with
and laboratory abnormalities of any cause are or without subsequent dose reduction. Adverse
summarized in Table 3. The most common ad- events of grade 3 or higher were observed in 58%
verse events in the triplet-therapy group were of patients in the triplet-therapy group, in 50% in
gastrointestinal-related and skin-related events, in- the doublet-therapy group, and in 61% in the
cluding diarrhea, nausea, vomiting, and acneiform control group. Discontinuation of therapy pri-
dermatitis. Low hemoglobin level or anemia was marily because of an adverse event was seen in
a common laboratory abnormality. Table S3 lists 7% of patients in the triplet-therapy group, in 8%
selected adverse events that occurred during the in the doublet-therapy group, and in 11% in the

1638 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

A Triplet-Therapy Group
Best Percentage Change from Baseline 80
60
40
20
0 *** ****

−20
−40
−60
−80
−100

B Doublet-Therapy Group
80
Best Percentage Change from Baseline

60
40
20
0 **

−20
−40
−60
−80
−100

C Control Group
80
Best Percentage Change from Baseline

60
40
20
0 *** ** *** * * ** ** *

−20
−40
−60
−80
−100

Figure 2. Best Percentage Change in Size of Target Lesions.

Shown are the best percentage changes from baseline in the sum of the diameters of the target lesions in each pa-
tient in the three groups, as determined by central review. The dashed lines at 20% and −30% indicate progressive
disease and partial response, respectively, according to Response Evaluation Criteria in Solid Tumors, version 1.1.
The asterisks indicate patients who had a complete response, partial response, or stable disease with respect to tar-
get lesions but who had a new lesion, a progressing nontarget lesion, or both.

control group. Fatal adverse events occurred in anaphylaxis, and one (in the control group) was
4%, 3%, and 4% of the patients, respectively. from respiratory failure.
Three of the deaths were determined by the in- The median duration of exposure to trial drugs
vestigators to be related to treatment: one death was 21 weeks in the triplet-therapy group, 19
(in the triplet-therapy group) was from colonic weeks in the doublet-therapy group, and 7 weeks
perforation, one (in the control group) was from in the control group. The median relative dose

n engl j med 381;17 nejm.org October 24, 2019 1639

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 3. Adverse Events and Selected Laboratory Abnormalities.*

Triplet Regimen Doublet Regimen Control

Variable (N = 222) (N = 216) (N = 193)

Any Grade Grade ≥3 Any Grade Grade ≥3 Any Grade Grade ≥3

number of patients (percent)

Adverse events
Any adverse event 217 (98) 128 (58) 212 (98) 108 (50) 188 (97) 117 (61)
Diarrhea 137 (62) 22 (10) 72 (33) 4 (2) 93 (48) 19 (10)
Acneiform dermatitis 108 (49) 5 (2) 63 (29) 1 (<1) 76 (39) 5 (3)
Nausea 100 (45) 10 (5) 74 (34) 1 (<1) 80 (41) 2 (1)
Vomiting 85 (38) 9 (4) 46 (21) 3 (1) 56 (29) 5 (3)
Fatigue 73 (33) 5 (2) 65 (30) 9 (4) 53 (27) 8 (4)
Abdominal pain 65 (29) 13 (6) 49 (23) 5 (2) 48 (25) 9 (5)
Decreased appetite 63 (28) 4 (2) 58 (27) 3 (1) 52 (27) 6 (3)
Asthenia 55 (25) 7 (3) 46 (21) 7 (3) 49 (25) 9 (5)
Constipation 55 (25) 0 33 (15) 0 35 (18) 2 (1)
Dry skin 46 (21) 2 (1) 24 (11) 0 13 (7) 1 (1)
Pyrexia 45 (20) 4 (2) 35 (16) 2 (1) 27 (14) 1 (1)
Rash 42 (19) 1 (<1) 25 (12) 0 27 (14) 3 (2)
Stomatitis 31 (14) 1 (<1) 12 (6) 0 44 (23) 4 (2)
Palmar–plantar erythrodysesthesia 28 (13) 0 9 (4) 1 (<1) 14 (7) 0
syndrome
Pruritus 28 (13) 0 20 (9) 0 9 (5) 0
Back pain 25 (11) 2 (1) 22 (10) 2 (1) 23 (12) 2 (1)
Blurred vision 25 (11) 0 8 (4) 0 1 (1) 0
Peripheral edema 24 (11) 1 (<1) 18 (8) 0 13 (7) 1 (1)
Weight decreased 24 (11) 1 (<1) 21 (10) 1 (<1) 11 (6) 0
Arthralgia 23 (10) 0 41 (19) 2 (1) 1 (1) 0
Cough 23 (10) 0 16 (7) 1 (<1) 10 (5) 0
Myalgia 18 (8) 0 29 (13) 1 (<1) 4 (2) 0
Dyspnea 17 (8) 2 (1) 23 (11) 2 (1) 17 (9) 5 (3)
Headache 16 (7) 0 42 (19) 0 5 (3) 0
Pain in extremity 15 (7) 0 22 (10) 0 1 (1) 0
Insomnia 11 (5) 0 24 (11) 0 11 (6) 0
Musculoskeletal pain 6 (3) 0 27 (12) 0 3 (2) 0
Melanocytic nevus 1 (<1) 0 31 (14) 0 0 0
Abnormal laboratory values
Alanine aminotransferase 51 (23) 4 (2) 36 (17) 0 50 (26) 5 (3)
Aspartate aminotransferase 50 (23) 4 (2) 31 (14) 3 (1) 38 (20) 3 (2)
Bilirubin 12 (5) 5 (2) 16 (7) 5 (2) 16 (8) 6 (3)
Creatine kinase 52 (23) 6 (3) 6 (3) 0 13 (7) 0
Creatinine 166 (75) 10 (5) 109 (50) 5 (2) 65 (34) 2 (1)
Hemoglobin 125 (56) 24 (11) 70 (32) 9 (4) 85 (44) 8 (4)

* Shown are adverse events of any grade and selected laboratory abnormalities reported in more than 10% of patients and adverse events of
grade 3 or higher reported in more than 2% of patients in the triplet-therapy group or the doublet-therapy group. Adverse events were grad-
ed according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03, and were coded according to
the preferred terms from the Medical Dictionary for Regulatory Activities, version 21.0.

1640 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

intensities were 91% in the triplet-therapy group survival that compared the triplet regimen with
and 98% in the doublet-therapy group for encoraf­ the doublet regimen showed a hazard ratio for
enib; 87% in the triplet-therapy group for binimeti­ death that favored the triplet regimen (0.79; 95%
nib; and 91% in the triplet-therapy group and CI, 0.59 to 1.06). A waterfall plot of the best
93% in the doublet-therapy group for cetuximab. percentage change in the sum of the diameters
of the target lesions also indicated differences in
Discussion the depth of response that favored the triplet
regimen (Fig. S3).
This initial analysis of the BEACON CRC trial The rate of adverse events was similar in the
shows that the triplet regimen of encorafenib, triplet-therapy group and the doublet-therapy
binimetinib, and cetuximab resulted in longer group, and the frequency of toxic effects of grade 3
overall survival and a higher objective response or higher was slightly higher in the control
rate than a control regimen of cetuximab plus the group than in either targeted-therapy group. The
investigators’ choice of irinotecan-based chemo- median duration of exposure to trial treatment
therapy in patients with BRAF V600E–mutated was 21 weeks in the triplet-therapy group, 19 weeks
metastatic colorectal cancer who had had dis- in the doublet-therapy group, and 7 weeks in the
ease progression after one or two previous regi- control group. The inclusion of binimetinib as
mens. The doublet regimen of encorafenib and part of the triplet regimen added some addi-
cetuximab also resulted in significantly longer tional toxic effects that were associated with
overall survival and a higher objective response MEK inhibition. The rate of treatment discon-
rate than were seen in the control group. The tinuation and the relative dose intensity were
results in the control group were as expected on similar in the triplet-therapy group and the
the basis of recently reported prospective data in doublet-therapy group. The observed safety pro-
a similar population and several published retro- files were generally consistent with those previ-
spective analyses.1,12,13,32,34 ously reported for each regimen and with the
The triplet regimen was designed to be a com- known effects of MEK, BRAF, and EGFR inhibi-
bination of agents that would provide the most tors. Headache, musculoskeletal pain, arthralgia,
effective inhibition of the MAPK pathway. Pre- and myalgia occurred more frequently in the
clinical and clinical studies showed that the lack doublet-therapy group than in the triplet-therapy
of efficacy of single-agent BRAF or dual BRAF group but did not generally result in discontinu-
and MEK inhibition in BRAF V600E–mutated ation of either component of the regimen. The
colorectal cancer is related to EGFR-mediated less frequent occurrence of these adverse effects
adaptive feedback — a finding that led to the in the triplet-therapy group is consistent with
development of a combination of BRAF, MEK, the ability of MEK inhibition to mitigate some
and EGFR inhibition through a series of iterative toxic effects associated with BRAF inhibition.15,16
studies.17-23,25 Analyses of mechanisms of resis- This initial analysis in patients with BRAF
tance in several studies reiterated the depen- V600E–mutated metastatic colorectal cancer who
dency of BRAF V600E–mutated metastatic colorec- had had disease progression after one or two
tal cancer on the MAPK pathway, which is previous regimens showed that a triplet regimen
composed most prominently of KRAS, NRAS, of encorafenib, binimetinib, and cetuximab or a
BRAF, and MEK.17,21,22,35,36 Suppression of MAPK doublet regimen of encorafenib and cetuximab,
signaling with encorafenib, binimetinib, and as compared with current standard therapy, re-
cetuximab represented the logical therapeutic sulted in a significant and clinically relevant
strategy to address this persistent dependency. benefit with respect to overall survival and ob-
Despite the documented activity of the triplet jective response rate. The side-effect profiles of
regimen, resistance ultimately develops in many both combination regimens allowed maintenance
patients, and further characterization of these of high dose intensity for the majority of pa-
mechanisms is needed to further improve out- tients and are consistent with the known profile
comes. of each agent. Further follow-up is needed to
The trial was not powered to compare the better define the relative benefits of the triplet
two experimental groups directly, and such a and doublet regimens.
comparison is further limited by the interim Supported by Array BioPharma, Merck (for sites outside North
nature of this analysis. The analysis of overall America), ONO Pharmaceutical, and Pierre Fabre.

n engl j med 381;17 nejm.org October 24, 2019 1641

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Dr. Kopetz reports receiving advisory board fees from Amal Pharmaceuticals; Dr. Loupakis, receiving lecture fees from
Therapeutics, Amgen, AstraZeneca, Bayer Healthcare, Boeh- Amgen and F. Hoffmann–La Roche, and advisory fees from
ringer Ingelheim, Boston Biomedical, Eli Lilly, EMD Serono, Bayer; Dr. Lonardi, receiving advisory board fees from Amgen,
Holy Stone Healthcare, Karyopharm Therapeutics, Merck, No- Eli Lilly, and Merck; Dr. Elez, receiving lecture fees and advisory
vartis, Roche, and Symphogen, grant support, paid to M.D. fees from Amgen, Merck, Pierre Fabre Pharmaceuticals, Sanofi
Anderson Cancer Center, from AstraZeneca, holding stock op- Aventis, and Servier, and advisory board fees from F. Hoff-
tions in Navire Pharma, and receiving lecture fees from Pierre mann–La Roche; Dr. Schellens, receiving consulting fees from
Fabre Pharmaceuticals; Dr. Grothey, receiving consulting fees, Debiopharm and being employed by and holding stock in Modra
paid to his institution, from Array BioPharma, Bayer Healthcare, Pharmaceuticals; Dr. Dekervel, receiving lecture fees from Am-
Boston Biomedical, Genentech, and Pierre Fabre Pharmaceuti- gen, Bayer, Ipsen Biopharmaceuticals, and Merck, travel support
cals; Dr. Yaeger, receiving grant support, paid to her institution, from Ipsen Biopharmaceuticals, and consulting fees from No-
from Array BioPharma, Boehringer Ingelheim, GlaxoSmith- vartis; Dr. Morris, receiving consulting fees from Array Bio-
Kline, and Novartis, and fees for serving on a steering commit- Pharma; Dr. Gollerkeri, being employed by Array BioPharma;
tee and travel support from Array BioPharma; Dr. Van Cutsem, Dr. Keir, being employed by Array BioPharma and Novartis
receiving grant support, paid to the University of Leuven, from Pharma; Dr. Maharry, being employed by Array BioPharma; Dr.
Amgen and Boehringer Ingelheim, and advisory board fees from Pickard, being employed by and holding stock in Array Bio-
Array BioPharma, AstraZeneca, Bayer Healthcare, Bristol-Myers Pharma; Dr. Christy-Bittel, being employed by Array BioPharma
Squibb, Celgene, Daiichi Sankyo, Eli Lilly, F. Hoffmann–La and holding stock in Gilead Sciences; Dr. Sandor, being em-
Roche, Incyte Corporation, Ipsen Biopharmaceuticals, Merck, ployed by and holding stock in Array BioPharma; and Dr. Tab-
Novartis, Pierre Fabre Pharmaceuticals, and Sirtex Medical; Dr. ernero, receiving consulting fees from Array BioPharma, As-
Desai, receiving fees for institutional collaboration, paid to his traZeneca, Bayer, BeiGene, Biocartis, Boehringer Ingelheim,
institution, from AstraZeneca, Bristol-Myers Squibb, Genentech Chugai Pharmaceutical, Eli Lilly and Company, F. Hoffmann–La
USA, and GlaxoSmithKline, and consulting fees from BeiGene; Roche, Foundation Medicine, Genentech, Genmab, HalioDX
Dr. Yoshino, receiving grant support, paid to National Cancer SAS, Halozyme, Imugene, Inflection Biosciences, Ipsen Biosci-
Center Hospital East, from Chugai Pharmaceutical, Daiichi ences, Kura Oncology, Menarini Ricerche, Merck Serono, Merri­
Sankyo, GlaxoSmithKline, MSD KK, Novartis Pharma, ONO mack Pharmaceuticals, Merus, Molecular Partners, MSD, Novar-
Pharmaceutical, Parexel International, Sanofi, and Sumitomo tis, Peptomyc, Pfizer, Pharmacyclics (an AbbVie company),
Dainippon Pharma; Dr. Wasan, receiving fees for serving on an ProteoDesign, Rafael Pharmaceuticals, Roche Diagnostics,
end-point review committee from Array BioPharma, travel sup- Sanofi, SeaGen, Seattle Genetics, Servier, Symphogen, Taiho
port from Bristol-Myers Squibb, advisory board fees and fees for Pharmaceutical, and VCN Biosciences. No other potential con-
serving on a speaker’s bureau from EMD Serono, F. Hoffmann– flict of interest relevant to this article was reported.
La Roche, and Servier, advisory board fees from Erytech, Incyte Disclosure forms provided by the authors are available with
Corporation, and Shire, and grant support, paid to his institu- the full text of this article at NEJM.org.
tion, and advisory board fees and fees for serving on a speaker’s A data sharing statement provided by the authors is available
bureau from Sirtex Medical; Dr. Ciardiello, receiving grant sup- with the full text of this article at NEJM.org.
port, paid to his institution, from Amgen, AstraZeneca, Bayer We thank the patients and their families, as well as the partici-
Healthcare, F. Hoffmann–La Roche, and Merck KGaA, and pating trial teams, for making this trial possible and J.D. Cox,
advisory board fees from Array BioPharma, Bayer Healthcare, Ph.D. (Mayville Medical Communications), for medical writing
F. Hoffmann–La Roche, Merck KGaA, Pfizer, and Pierre Fabre and editorial assistance with an earlier version of the manuscript.

Appendix
The authors’ full names and academic degrees are as follows: Scott Kopetz, M.D., Ph.D., Axel Grothey, M.D., Rona Yaeger, M.D., Eric
Van Cutsem, M.D., Ph.D., Jayesh Desai, M.B., B.S., Takayuki Yoshino, M.D., Ph.D., Harpreet Wasan, M.D., Fortunato Ciardiello, M.D.,
Ph.D., Fotios Loupakis, M.D., Ph.D., Yong Sang Hong, M.D., Ph.D., Neeltje Steeghs, M.D., Ph.D., Tormod K. Guren, M.D., Ph.D.,
Hendrik‑Tobias Arkenau, M.D., Ph.D., Pilar Garcia‑Alfonso, M.D., Per Pfeiffer, M.D., Ph.D., Sergey Orlov, M.D., Ph.D., Sara Lonardi,
M.D., Elena Elez, M.D., Ph.D., Tae‑Won Kim, M.D., Ph.D., Jan H.M. Schellens, M.D., Ph.D., Christina Guo, M.B., B.S., Asha Krishnan,
B.S., Jeroen Dekervel, M.D., Van Morris, M.D., Aitana Calvo Ferrandiz, M.D., Ph.D., L.S. Tarpgaard, M.D., Ph.D., Michael Braun, M.B.,
Ch.B., Ph.D., Ashwin Gollerkeri, M.D., Christopher Keir, M.D., Kati Maharry, Ph.D., Michael Pickard, Ph.D., Janna Christy‑Bittel,
M.S.N., Lisa Anderson, M.P.H., Victor Sandor, M.D., and Josep Tabernero, M.D., Ph.D.
The authors’ affiliations are as follows: the University of Texas M.D. Anderson Cancer Center, Houston (S.K., V.M.); West Cancer
Center and Research Institute, OneOncology, Germantown, TN (A. Grothey); Memorial Sloan Kettering Cancer Center, New York (R.Y.,
A.K.); University Hospital Gasthuisberg and University of Leuven, Leuven, Belgium (E.V.C., J. Dekervel); the Peter MacCallum Cancer
Centre, Melbourne, VIC, Australia (J. Desai, C.G.); National Cancer Center Hospital East, Kashiwa, Japan (T.Y.); Hammersmith Hospi-
tal, Division of Cancer, Imperial College London (H.W.), and the Sarah Cannon Research Institute and University College London
Cancer Institute (H.-T.A.), London, and the Christie NHS Foundation Trust/National Institute for Health Research Manchester Bio-
medical Research Centre, Manchester (M.B.) — all in the United Kingdom; the University of Campania Luigi Vanvitelli, Naples (F.C.),
and Istituto Oncologico Veneto, Istituto di Ricovero e Cura a Carattere Scientifico, Padua (F.L., S.L.) — both in Italy; Asan Medical
Center, University of Ulsan College of Medicine, Seoul, South Korea (Y.S.H., T.-W.K.); the Netherlands Cancer Institute, Amsterdam
(N.S., J.H.M.S.); Oslo University Hospital, Oslo (T.K.G.); Hospital Gregorio Marañón, Madrid (P.G.-A., A.C.F.), and Vall d’Hebron
University Hospital, Vall d’Hebron Institute of Oncology, UVic, IOB-Quiron, Barcelona (E.E., J.T.) — both in Spain; Odense University
Hospital, Odense, Denmark (P.P., L.S.T.); Pavlov First St. Petersburg State Medical University, St. Petersburg, Russia (S.O.); and Array
BioPharma, Boulder, CO (A. Gollerkeri, C.K., K.M., M.P., J.C.-B., L.A., V.S.).

1642 n engl j med 381;17 nejm.org October 24, 2019

The New England Journal of Medicine

Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.
 BR AF V600E–Mutated Colorectal Cancer

References
1. Davies H, Bignell GR, Cox C, et al. et al. Dabrafenib plus trametinib in patients inhibition with vemurafenib. Cancer Dis-
Mutations of the BRAF gene in human with previously treated BRAF(V600E)- cov 2012;​2:​227-35.
cancer. Nature 2002;​417:​949-54. mutant metastatic non-small cell lung 26. Delord JP, Robert C, Nyakas M, et al.
2. De Roock W, Claes B, Bernasconi D, cancer: an open-label, multicentre phase 2 Phase I dose-escalation and -expansion
et al. Effects of KRAS, BRAF, NRAS, and trial. Lancet Oncol 2016;​17:​984-93. study of the braf inhibitor encorafenib
PIK3CA mutations on the efficacy of 15. Dummer R, Ascierto PA, Gogas HJ, (LGX818) in metastatic BRAF-mutant mela-
­cetuximab plus chemotherapy in chemo- et al. Encorafenib plus binimetinib versus noma. Clin Cancer Res 2017;​23:​5339-48.
therapy-refractory metastatic colorectal vemurafenib or encorafenib in patients with 27. Tabernero J, Van Geel R, Guren TK,
cancer: a retrospective consortium analy- BRAF-mutant melanoma (COLUMBUS): et al. Phase 2 results: encorafenib (ENCO)
sis. Lancet Oncol 2010;​11:​753-62. a multicentre, open-label, randomised and cetuximab (CETUX) with or without
3. Sorbye H, Dragomir A, Sundström M, phase 3 trial. Lancet Oncol 2018;​19:​603- alpelisib (ALP) in patients with advanced
et al. High BRAF mutation frequency and 15. BRAF-mutant colorectal cancer (BRAFm
marked survival differences in subgroups 16. Dummer R, Ascierto PA, Gogas HJ, CRC). J Clin Oncol 2016;​34:​Suppl:​3544.
according to KRAS/BRAF mutation status et al. Overall survival in patients with abstract.
and tumor tissue availability in a prospec- BRAF-mutant melanoma receiving en­ 28. Van Cutsem E, Huijberts S, Grothey A,
tive population-based metastatic colorec- corafenib plus binimetinib versus vemu- et al. Binimetinib, encorafenib, and cetuxi­
tal cancer cohort. PLoS One 2015;​10(6):​ rafenib or encorafenib (COLUMBUS): mab triplet therapy for patients with
e0131046. a multicentre, open-label, randomised, BRAF V600E–mutant metastatic colorec-
4. Loupakis F, Ruzzo A, Cremolini C, phase 3 trial. Lancet Oncol 2018;​19:​1315- tal cancer: safety lead-in results from the
et al. KRAS codon 61, 146 and BRAF muta- 27. phase III BEACON colorectal cancer study.
tions predict resistance to cetuximab plus 17. Kopetz S, Desai J, Chan E, et al. Phase J Clin Oncol 2019;​37:​1460-9.
irinotecan in KRAS codon 12 and 13 wild- II pilot study of vemurafenib in patients 29. Eisenhauer EA, Therasse P, Bogaerts J,
type metastatic colorectal cancer. Br J Can- with metastatic BRAF-mutated colorectal et al. New response evaluation criteria in
cer 2009;​101:​715-21. cancer. J Clin Oncol 2015;​33:​4032-8. solid tumours: revised RECIST guideline
5. AACR Project GENIE Consortium. 18. Corcoran RB, Atreya CE, Falchook GS, (version 1.1). Eur J Cancer 2009;​45:​228-47.
AACR Project GENIE: powering precision et al. Combined BRAF and MEK inhibi- 30. Department of Health and Human
medicine through an international con- tion with dabrafenib and trametinib in Services. Common terminology criteria
sortium. Cancer Discov 2017;​7:​818-31. BRAF V600-mutant colorectal cancer. J Clin for adverse events, version 4.0. (CTCAE).
6. Cancer Genome Atlas Research Net- Oncol 2015;​33:​4023-31. May 28, 2009 (https://evs​.nci​.nih​.gov/​f tp1/​
work. Comprehensive genomic character- 19. Prahallad A, Sun C, Huang S, et al. CTCAE/​CTCAE_4​.03/​A rchive/​CTCAE_4​.0
ization of squamous cell lung cancers. Unresponsiveness of colon cancer to _2009​-­05​-­29_QuickReference_8​.5x11​.pdf).
Nature 2012;​489:​519-25. BRAF(V600E) inhibition through feed- 31. Wiens BL, Dmitrienko A. The fallback
7. Yaeger R, Chatila WK, Lipsyc MD, et al. back activation of EGFR. Nature 2012;​ procedure for evaluating a single family
Clinical sequencing defines the genomic 483:​100-3. of hypotheses. J Biopharm Stat 2005;​15:​
landscape of metastatic colorectal cancer. 20. Mao M, Tian F, Mariadason JM, et al. 929-42.
Cancer Cell 2018;​33(1):​125.e3-136.e3. Resistance to BRAF inhibition in BRAF- 32. Morris V, Overman MJ, Jiang Z-Q, et al.
8. Barras D, Missiaglia E, Wirapati P, mutant colon cancer can be overcome Progression-free survival remains poor
et al. BRAF V600E mutant colorectal can- with PI3K inhibition or demethylating over sequential lines of systemic therapy
cer subtypes based on gene expression. agents. Clin Cancer Res 2013;​ 19:​657- in patients with BRAF-mutated colorectal
Clin Cancer Res 2017;​23:​104-15. 67. cancer. Clin Colorectal Cancer 2014;​13:​
9. Bylsma LC, Gillezeau C, Garawin T, 21. Corcoran RB, André T, Atreya CE, et al. 164-71.
et al. Prevalence of RAS and BRAF muta- Combined BRAF, EGFR, and MEK inhibi- 33. Chibaudel B, Bonnetain F, Shi Q, et al.
tions in metastatic colorectal cancer tion in patients with BRAFV600E-mutant Alternative end points to evaluate a thera-
(mCRC) patients by tumor location. J Clin colorectal cancer. Cancer Discov 2018;​8:​ peutic strategy in advanced colorectal
Oncol 2018;​36:​Suppl:​681. abstract. 428-43. cancer: evaluation of progression-free
10. Clarke CN, Kopetz ES. BRAF mutant 22. Hong DS, Morris VK, El Osta B, et al. survival, duration of disease control, and
colorectal cancer as a distinct subset of Phase IB study of vemurafenib in combi- time to failure of strategy — an Aide et Re-
colorectal cancer: clinical characteristics, nation with irinotecan and cetuximab in cherche en Cancerologie Digestive Group
clinical behavior, and response to target- patients with metastatic colorectal cancer study. J Clin Oncol 2011;​29:​4199-204.
ed therapies. J Gastrointest Oncol 2015;​6:​ with BRAFV600E mutation. Cancer Dis- 34. Peeters M, Price TJ, Cervantes A, et al.
660-7. cov 2016;​6:​1352-65. Final results from a randomized phase 3
11. Tran B, Kopetz S, Tie J, et al. Impact 23. Yaeger R, Cercek A, O’Reilly EM, et al. study of FOLFIRI {+/-} panitumumab for
of BRAF mutation and microsatellite in- Pilot trial of combined BRAF and EGFR second-line treatment of metastatic colo­
stability on the pattern of metastatic inhibition in BRAF-mutant metastatic rectal cancer. Ann Oncol 2014;​25:​107-16.
spread and prognosis in metastatic colorec- colorectal cancer patients. Clin Cancer 35. Ahronian LG, Sennott EM, Van Allen
tal cancer. Cancer 2011;​117:​4623-32. Res 2015;​21:​1313-20. EM, et al. Clinical acquired resistance to
12. Loupakis F, Cremolini C, Masi G, 24. Corcoran RB, Dias-Santagata D, Ber- RAF inhibitor combinations in BRAF-
et al. Initial therapy with FOLFOXIRI and gethon K, Iafrate AJ, Settleman J, Engel- mutant colorectal cancer through MAPK
bevacizumab for metastatic colorectal can- man JA. BRAF gene amplification can pathway alterations. Cancer Discov 2015;​
cer. N Engl J Med 2014;​371:​1609-18. promote acquired resistance to MEK in- 5:​358-67.
13. Kopetz S, McDonough SL, Lenz H-J, hibitors in cancer cells harboring the 36. Yaeger R, Yao Z, Hyman DM, et al.
et al. Randomized trial of irinotecan and BRAF V600E mutation. Sci Signal 2010;​ Mechanisms of acquired resistance to
cetuximab with or without vemurafenib 3(149):​ra84. BRAF V600E inhibition in colon cancers
in BRAF-mutant metastatic colorectal can- 25. Corcoran RB, Ebi H, Turke AB, et al. converge on RAF dimerization and are
cer (SWOG S1406). J Clin Oncol 2017;​35:​ EGFR-mediated re-activation of MAPK sensitive to its inhibition. Cancer Res 2017;​
Suppl:​3505. abstract. signaling contributes to insensitivity of 77:​6513-23.
14. Planchard D, Besse B, Groen HJM, BRAF mutant colorectal cancers to RAF Copyright © 2019 Massachusetts Medical Society.

n engl j med 381;17 nejm.org October 24, 2019 1643

The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2024. For personal use only.
No other uses without permission. Copyright © 2019 Massachusetts Medical Society. All rights reserved.